Abstract: As a large agricultural country, China has a large area of farmland coverage, which requires a lot of irrigation water, but the utilization rate of farmland irrigation water resources is not ideal. In the agricultural irrigation operation, the irrigation system is not intelligent and automatic enough, which not only cannot help the staff to complete the irrigation operation efficiently, but also causes a serious waste of water resources, which to some extent hinders the agricultural development in China. Based on the above phenomena, this paper proposes an intelligent irrigation decision-making system based on big data, which adopts convolutional neural network and fuzzy control to predict and control the irrigation water consumption, and designs an intelligent irrigation system. The experimental results show that the algorithm can correctly predict the farmland water consumption and achieve the purpose of saving water resources.